cicd-automation-workflow-automate
Pass
Audited by Gen Agent Trust Hub on Jul 7, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it incorporates untrusted user requirements via the
$ARGUMENTSvariable inSKILL.md. An attacker could embed malicious instructions within these requirements to influence the agent's behavior during workflow generation. - Ingestion points:
SKILL.md(via$ARGUMENTSplaceholder). - Boundary markers: None present.
- Capability inventory: The skill can generate and propose the execution of shell scripts (
scripts/setup-dev-environment.sh), CI/CD configuration files (GitHub Actions, GitLab CI), and infrastructure-as-code (Terraform). - Sanitization: No evidence of input validation or sanitization for the provided arguments.
- [EXTERNAL_DOWNLOADS]: The skill's implementation playbook references numerous third-party GitHub Actions, Node.js packages, and Python packages for CI/CD, security scanning, and development environment setup. These include official actions from trusted organizations such as GitHub, AWS, Snyk, and Aqua Security, as well as well-known community tools. It also configures Helm repositories for official Prometheus and Grafana charts.
Audit Metadata